HI Larry -
I added a comparison of various shutter speeds to the original post:
http://www.markcassino.com/b2evolution/index.php/pentax-k-1-banding?blog=9
Obviously, I had to stop down quite a bit to shoot at the slow shutter
speeds, so some of the detail in the wax paper diffuser becomes apparent
in the slow shutter speeds, despite the lens being defocused.
This only happens with the electronic shutter. I saw the same thing with
the Q when trying to photograph insects. See the bee photos in this old
post:
http://www.markcassino.com/b2evolution/index.php/more-pentax-q-macro-experiments-1
There is also a Wikipedia article on it.
In my tests I saw no banding with the mechanical shutter.
I found a brand of bulb that seems to cause almost no banding and
started using those today. Actually fairly inexpensive bulbs sold at
Walgreens.
Mark
On 3/4/2017 3:40 PM, Larry Colen wrote:
Mark C wrote:
I did that and will put together the results. I saw falloff between
shutter speeds - 1/45, 1/90, and 1/180 had more pronounced banding that
1/60 and 1/125. But all those speeds had some banding. All speeds <1/45
showed no banding and all > 1/180 showed about the same degree of
banding. I'll label the images and put them into a single file.
Very cool. This was with the electronic shutter if I recall
correctly? Did the number of bands change with different shutter
speeds? I'm wondering if the length of time to expose the full sensor
changes, and they just expose each row for different amounts of time.
Did you get different results with electronic versus mechanical shutters?
Mark
On 3/2/2017 4:26 PM, Larry Colen wrote:
<signal processing geek>
Mark, as an experiment, please try this at various multiples of 1/120
second (83mSec):
1/120, 2/120, 3/120, 4/120, 5/120 -> 1/120, 1/60, 1/40, 1/30, 1/24,
1/20
Since the light is effectively a rectified sine wave, any time you
integrate (expose) over a half power cycle (1/20), you should get
exactly a full cycle of the light, and it should come out even with no
banding.
If you do run an experiment where you just cycle through exposure
times from, for example, 1/500 down to 1/5, at 1/3 stop (1dB) and post
the results, that should be a very cool demonstration of
aliasing/nyquist rate etc. and I've got some science teacher friends
who'd probably really appreciate that gallery being up on the web.
</signal processing geek>
--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow
the directions.
